DESCRIPTION: (Applicant's Abstract) Integration or linkage of ambulatory substance abuse and medical treatment promises to improve health care utilization and clinical outcomes of illicit drug users but data supporting this benefit are limited. The proposed project will evaluate the association of linked and unlinked substance abuse and medical care with diverse outcomes of HIV infected (HIV+) and HIV uninfected (HIV-) drug users. The investigators will employ population-based, administrative data for 40,000 drug users in New York State (NYS) Medicaid combined with interview data from substance abuse and medical clinical directors about specific features of clinic area. A case study will also be conducted using detailed, patient-level data for 600 current or previous enrollees in Montefiore Center's linked substance abuse and medical are program. The specific aims are: 1) To examine associations of the presence and types of substance abuse and medical care linkages with 4 outcome categories: utilization (i.e., hospitalization, ER visits dental care, HIV related therapies/services, HIV testing), substance abuse (i.e., abuse related complications, retention in drug treatment), other medical (i.e., Pneumocystis carinii pneumonia, other pneumonias, mortality), and financial (i.e., Medicaid expenditures). 2) To examine associations of specific clinic features (i.e., staffing, services, structure) with aim #1 outcomes. 3) To examine the association of detailed patient characteristics with 3 outcome categories: utilization (i.e., hospitalization, ER visits), substance abuse (i.e., abstinence, abuse-related complications), and other medical (i.e., progression of HIV disease). For aim #1, Medicaid files for 1995-7 will be analyzed for 20,000 persons from among Medicaid drug users with no such longitudinal care in 1995 and a comparison group, selected randomly after stratification by HIV status, of 20,000 persons from among Medicaid drug users with no such longitudinal care in 1995. For aim #2, specific aspects of services, staffing, and delivery of care will be obtained from telephone interviews of directors of 125 randomly selected substance abuse treatment clinics (of 255) and 350 other clinics serving as usual sources of medical care in 1995 for the Medicaid study sample. Provider interview and patient claims files will be linked for analysis. For aim #3, chart, patient interview, and Medicaid records will be studied for 600 drug users followed at Montefiore to obtain a comprehensive view of their care. The large, population-based sample will yield empirical data on optimal models of care for this medically complex population. Health planners and providers require these data to foster and develop efficient systems for care for drug users in a changing health care environment.